This application is in response to PA-09-262: NIDDK: Health Disparities in NIDDK Diseases. Ineffective diabetes care impacts patient well-being, health outcomes, health care costs and contributes to racial and ethnic disparities. Latinos have high rates of diabetes and well-documented disparities in diabetes care and outcomes. Factors recognized as contributing to disparities in diabetes outcomes for Latinos include lack of health insurance, financial barriers to accessing treatments, and limited English proficiency. Latinos with limited English proficiency (LEP) now constitute 44% of Latino adults. Health care systems are struggling to provide clinical care to Latinos with LEP and diabetes using either bilingual clinicians or interpreter services. The comparative effectiveness of these language access strategies with respect to clinical outcomes is not known. In a prior study, in a setting with uniform access to care and professional medical interpreters, we found that Latino patients with LEP were less likely to have poor glycemic control if they identified their physician as fluent in Spanish. We now propose to build on our work by examining multiple clinical outcomes using richly detailed data from a 5-year longitudinal cohort of Latinos enrolled in the Kaiser Permanente Northern California-Diabetes Registry. The overarching goals of this proposal are: to determine whether language proficiency is a root cause of diabetes disparities among Latinos; provide essential information on the comparative effectiveness of two strategies to overcome language barriers (bilingual clinicians versus professional medical interpreters); deepen understanding of how language barriers influence patient and physician behavior (medication adherence and clinical inertia); and determine if there are aspects of interpreter use that can be modified to improve the outcomes of language discordant care. Our results will inform health systems and clinicians seeking to improve diabetes care for Latinos with LEP and will provide a framework for intervention studies to reduce disparities for this population.